Introduction To Randomized Controlled Clinical Trials Second Edition Texts In Statistical Science Series 2nd Edition John Ns Matthews by John N.s. Matthews 9781584886242, 1584886242 instant download after payment.

Prefaces

First Edition

Clinical trials are experiments performed on human subjects, usually

patients, in order to assess the efficacy of a treatment that is under investigation.

Over the last two to three decades randomized concurrently controlled

clinical trials have become established as the method which

investigators must use to assess new treatments if their claims are to find

widespread acceptance. The methodology underpinning these trials is firmly

based in statistical theory, and the success of randomized trials perhaps

constitutes the greatest achievement of statistics in the second half of the

twentieth century. As such it is important that students of statistics should

be able to study this area of their subject as soon as possible in their courses.

Whereas there are many excellent books on clinical trial methodology,

almost all are written for the practitioner, whether statistician or doctor, who

is about to participate in the running of a trial. There is a natural tendency

in such books both to cover administrative issues and to keep to a minimum

any mathematical passages. However, while trial administration is of

undoubted importance, too much emphasis on it is unnecessary and an

unwelcome distraction for students making their first acquaintance with the

underlying principles. Moreover, for a readership whose knowledge of mathematics

is likely to be substantially greater than their knowledge of medicine,

many of the principles involved can be introduced more precisely and succinctly

by the appropriate use of mathematics. This book is intended as an

introduction to the statistical methodology that underpins randomized controlled

trials, and is aimed primarily at the student of statistics. Administrative

aspects receive little emphasis and, if it is believed that it would help

the primary readership, a mathematical approach is adopted.

Although it is hoped that many will find the book useful as an introduction

to the subject, the needs of final-year undergraduate or postgraduate students

at British universities have been my main concern. However, there is

considerable variation within this group in the statistical techniques they

already know, and I have attempted to rely only on a few basic prerequisites.

This has led to a slight distortion of the subject matter. For example, so that

a knowledge of logistic regression is not needed for this book, trials with

binary outcomes are given less prominence than might be expected from

their prevalence in practice. The complete avoidance of survival analysis

perhaps leads to an even greater distortion. However, I believe these distortions

are justified by the wider accessibility that results.

Second Edition

The change of publisher between editions, from Edward Arnold to Chapman

& Hall/CRC, has given me the opportunity to make corrections and add

some new material. The aim, to write a book for students of statistics focusing

on statistical rather than administrative aspects of clinical trials, remains

the same. Some extra material on more sophisticated methods for balancing

clinical trials with binary or survival time outcomes. Many trials have outcomes

of this kind, and in the preface to the first edition I acknowledged

that the exclusion of outcomes of these forms was a distortion of the subject.

This was done to ensure that the book did not require the reader to be

familiar with more sophisticated methods of analysis, such as logistic regression

and proportional hazards models. On reflection, this was an unnecessary

restriction. None of the chapters in the book depend on Chapter 7, so

those wishing to avoid this material can still avoid it. The chapter is quite

long because an attempt has been made to provide brief introductions to

the required techniques.